parser.c

早期freebsd实现

				USTPUTC(c, out);
				break;
			case CRP:	/* ')' in arithmetic */
				if (parenlevel > 0) {
					USTPUTC(c, out);
					--parenlevel;
				} else {
					if (pgetc() == ')') {
						if (--arinest == 0) {
							USTPUTC(CTLENDARI, out);
							syntax = prevsyntax;
						} else
							USTPUTC(')', out);
					} else {
						/* 
						 * unbalanced parens
						 *  (don't 2nd guess - no error)
						 */
						pungetc();
						USTPUTC(')', out);
					}
				}
				break;
			case CBQUOTE:	/* '`' */
				PARSEBACKQOLD();
				break;
			case CEOF:
				goto endword;		/* exit outer loop */
			default:
				if (varnest == 0)
					goto endword;	/* exit outer loop */
				USTPUTC(c, out);
			}
			c = pgetc_macro();
		}
	}
endword:
	if (syntax == ARISYNTAX)
		synerror("Missing '))'");
	if (syntax != BASESYNTAX && ! parsebackquote && eofmark == NULL)
		synerror("Unterminated quoted string");
	if (varnest != 0) {
		startlinno = plinno;
		synerror("Missing '}'");
	}
	USTPUTC('\0', out);
	len = out - stackblock();
	out = stackblock();
	if (eofmark == NULL) {
		if ((c == '>' || c == '<')
		 && quotef == 0
		 && len <= 2
		 && (*out == '\0' || is_digit(*out))) {
			PARSEREDIR();
			return lasttoken = TREDIR;
		} else {
			pungetc();
		}
	}
	quoteflag = quotef;
	backquotelist = bqlist;
	grabstackblock(len);
	wordtext = out;
	return lasttoken = TWORD;
/* end of readtoken routine */



/*
 * Check to see whether we are at the end of the here document.  When this
 * is called, c is set to the first character of the next input line.  If
 * we are at the end of the here document, this routine sets the c to PEOF.
 */

checkend: {
	if (eofmark) {
		if (striptabs) {
			while (c == '\t')
				c = pgetc();
		}
		if (c == *eofmark) {
			if (pfgets(line, sizeof line) != NULL) {
				register char *p, *q;

				p = line;
				for (q = eofmark + 1 ; *q && *p == *q ; p++, q++);
				if (*p == '\n' && *q == '\0') {
					c = PEOF;
					plinno++;
					needprompt = doprompt;
				} else {
					pushstring(line, strlen(line), NULL);
				}
			}
		}
	}
	goto checkend_return;
}


/*
 * Parse a redirection operator.  The variable "out" points to a string
 * specifying the fd to be redirected.  The variable "c" contains the
 * first character of the redirection operator.
 */

parseredir: {
	char fd = *out;
	union node *np;

	np = (union node *)stalloc(sizeof (struct nfile));
	if (c == '>') {
		np->nfile.fd = 1;
		c = pgetc();
		if (c == '>')
			np->type = NAPPEND;
		else if (c == '&')
			np->type = NTOFD;
		else {
			np->type = NTO;
			pungetc();
		}
	} else {	/* c == '<' */
		np->nfile.fd = 0;
		c = pgetc();
		if (c == '<') {
			if (sizeof (struct nfile) != sizeof (struct nhere)) {
				np = (union node *)stalloc(sizeof (struct nhere));
				np->nfile.fd = 0;
			}
			np->type = NHERE;
			heredoc = (struct heredoc *)stalloc(sizeof (struct heredoc));
			heredoc->here = np;
			if ((c = pgetc()) == '-') {
				heredoc->striptabs = 1;
			} else {
				heredoc->striptabs = 0;
				pungetc();
			}
		} else if (c == '&')
			np->type = NFROMFD;
		else {
			np->type = NFROM;
			pungetc();
		}
	}
	if (fd != '\0')
		np->nfile.fd = digit_val(fd);
	redirnode = np;
	goto parseredir_return;
}


/*
 * Parse a substitution.  At this point, we have read the dollar sign
 * and nothing else.
 */

parsesub: {
	int subtype;
	int typeloc;
	int flags;
	char *p;
#ifndef GDB_HACK
	static const char types[] = "}-+?=";
#endif

	c = pgetc();
	if (c != '(' && c != '{' && !is_name(c) && !is_special(c)) {
		USTPUTC('$', out);
		pungetc();
	} else if (c == '(') {	/* $(command) or $((arith)) */
		if (pgetc() == '(') {
			PARSEARITH();
		} else {
			pungetc();
			PARSEBACKQNEW();
		}
	} else {
		USTPUTC(CTLVAR, out);
		typeloc = out - stackblock();
		USTPUTC(VSNORMAL, out);
		subtype = VSNORMAL;
		if (c == '{') {
			c = pgetc();
			subtype = 0;
		}
		if (is_name(c)) {
			do {
				STPUTC(c, out);
				c = pgetc();
			} while (is_in_name(c));
		} else {
			if (! is_special(c))
badsub:				synerror("Bad substitution");
			USTPUTC(c, out);
			c = pgetc();
		}
		STPUTC('=', out);
		flags = 0;
		if (subtype == 0) {
			if (c == ':') {
				flags = VSNUL;
				c = pgetc();
			}
			p = strchr(types, c);
			if (p == NULL)
				goto badsub;
			subtype = p - types + VSNORMAL;
		} else {
			pungetc();
		}
		if (dblquote || arinest)
			flags |= VSQUOTE;
		*(stackblock() + typeloc) = subtype | flags;
		if (subtype != VSNORMAL)
			varnest++;
	}
	goto parsesub_return;
}


/*
 * Called to parse command substitutions.  Newstyle is set if the command
 * is enclosed inside $(...); nlpp is a pointer to the head of the linked
 * list of commands (passed by reference), and savelen is the number of
 * characters on the top of the stack which must be preserved.
 */

parsebackq: {
	struct nodelist **nlpp;
	int savepbq;
	union node *n;
	char *volatile str;
	struct jmploc jmploc;
	struct jmploc *volatile savehandler;
	int savelen;

	savepbq = parsebackquote;
	if (setjmp(jmploc.loc)) {
		if (str)
			ckfree(str);
		parsebackquote = 0;
		handler = savehandler;
		longjmp(handler->loc, 1);
	}
	INTOFF;
	str = NULL;
	savelen = out - stackblock();
	if (savelen > 0) {
		str = ckmalloc(savelen);
		bcopy(stackblock(), str, savelen);
	}
	savehandler = handler;
	handler = &jmploc;
	INTON;
        if (oldstyle) {
                /* We must read until the closing backquote, giving special
                   treatment to some slashes, and then push the string and
                   reread it as input, interpreting it normally.  */
                register char *out;
                register c;
                int savelen;
                char *str;
 
                STARTSTACKSTR(out);
                while ((c = pgetc ()) != '`') {
                       if (c == '\\') {
                                c = pgetc ();
                                if (c != '\\' && c != '`' && c != '$'
                                    && (!dblquote || c != '"'))
                                        STPUTC('\\', out);
                       }
                       STPUTC(c, out);
                }
                STPUTC('\0', out);
                savelen = out - stackblock();
                if (savelen > 0) {
                        str = ckmalloc(savelen);
                        bcopy(stackblock(), str, savelen);
                }
                setinputstring(str, 1);
        }
	nlpp = &bqlist;
	while (*nlpp)
		nlpp = &(*nlpp)->next;
	*nlpp = (struct nodelist *)stalloc(sizeof (struct nodelist));
	(*nlpp)->next = NULL;
	parsebackquote = oldstyle;
	n = list(0);
        if (!oldstyle && (readtoken() != TRP))
                synexpect(TRP);
	(*nlpp)->n = n;
        /* Start reading from old file again.  */
        if (oldstyle)
                popfile();
	while (stackblocksize() <= savelen)
		growstackblock();
	STARTSTACKSTR(out);
	if (str) {
		bcopy(str, out, savelen);
		STADJUST(savelen, out);
		INTOFF;
		ckfree(str);
		str = NULL;
		INTON;
	}
	parsebackquote = savepbq;
	handler = savehandler;
	if (arinest || dblquote)
		USTPUTC(CTLBACKQ | CTLQUOTE, out);
	else
		USTPUTC(CTLBACKQ, out);
	if (oldstyle)
		goto parsebackq_oldreturn;
	else
		goto parsebackq_newreturn;
}

/*
 * Parse an arithmetic expansion (indicate start of one and set state)
 */
parsearith: {

	if (++arinest == 1) {
		prevsyntax = syntax;
		syntax = ARISYNTAX;
		USTPUTC(CTLARI, out);
	} else {
		/*
		 * we collapse embedded arithmetic expansion to
		 * parenthesis, which should be equivalent
		 */
		USTPUTC('(', out);
	}
	goto parsearith_return;
}

} /* end of readtoken */



#ifdef mkinit
RESET {
	tokpushback = 0;
	checkkwd = 0;
}
#endif

/*
 * Returns true if the text contains nothing to expand (no dollar signs
 * or backquotes).
 */

STATIC int
noexpand(text)
	char *text;
	{
	register char *p;
	register char c;

	p = text;
	while ((c = *p++) != '\0') {
		if (c == CTLESC)
			p++;
		else if (BASESYNTAX[c] == CCTL)
			return 0;
	}
	return 1;
}


/*
 * Return true if the argument is a legal variable name (a letter or
 * underscore followed by zero or more letters, underscores, and digits).
 */

int
goodname(name)
	char *name;
	{
	register char *p;

	p = name;
	if (! is_name(*p))
		return 0;
	while (*++p) {
		if (! is_in_name(*p))
			return 0;
	}
	return 1;
}


/*
 * Called when an unexpected token is read during the parse.  The argument
 * is the token that is expected, or -1 if more than one type of token can
 * occur at this point.
 */

STATIC void
synexpect(token) {
	char msg[64];

	if (token >= 0) {
		fmtstr(msg, 64, "%s unexpected (expecting %s)",
			tokname[lasttoken], tokname[token]);
	} else {
		fmtstr(msg, 64, "%s unexpected", tokname[lasttoken]);
	}
	synerror(msg);
}


STATIC void
synerror(msg)
	char *msg;
	{
	if (commandname)
		outfmt(&errout, "%s: %d: ", commandname, startlinno);
	outfmt(&errout, "Syntax error: %s\n", msg);
	error((char *)NULL);
}

STATIC void
setprompt(which)
	int which;
	{
	whichprompt = which;

	if (!el)
		out2str(getprompt(NULL));
}

/*
 * called by editline -- any expansions to the prompt
 *    should be added here.
 */
char *
getprompt(unused)
	void *unused;
	{
	switch (whichprompt) {
	case 0:
		return "";
	case 1:
		return ps1val();
	case 2:
		return ps2val();
	default:
		return "<internal prompt error>";
	}
}